Field of the Invention
The present invention relates to an image processing apparatus which generates a cross section image from three-dimensional image data, an image processing method, a medical image diagnostic system, and a storage medium.
Description of the Related Art
In the medical field, a doctor displays the medical image obtained by imaging an object (three-dimensional image data formed from a tomographic image group representing information of the inside of the object) on a monitor, and diagnoses a lesion portion by interpreting the displayed medical image. Medical image acquisition apparatuses (to be referred to as modalities hereinafter) which obtain three-dimensional image data include a magnetic resonance imaging apparatus (to be referred to as an MRI apparatus hereinafter), a computed tomography apparatus (to be referred to as a CT apparatus hereinafter), and a three-dimensional ultrasonic imaging apparatus.
More specifically, the doctor displays, on the monitor, a cross section image in a predetermined posture, which includes a target portion (to be referred to as a portion of interest hereinafter) such as a lesion portion in three-dimensional image data. The doctor then observes and diagnoses the portion of interest on the cross section image based on appearance features (to be referred to as image features hereinafter) of the portion of interest and its peripheral portion. As a cross section image in a predetermined posture, an image on an axial plane, sagittal plane, coronal plane, or a combination of them (three cross sections) is generally used.
In addition, Japanese Patent Laid-Open No. 2005-169070 discloses a technique of generating and displaying a cross section image including a designation point from three-dimensional image data by rotating the posture of a cross section in an arbitrary direction under a constraint condition that a cross section always includes a point designated in three-dimensional image data when changing the cross section. Furthermore, Japanese Patent Laid-Open No. 2011-125568 discloses a technique of rotating a posture around a straight line passing through two selected points of interest.
According to Japanese Patent Laid-Open Nos. 2005-169070 and 2011-125568, however, a cross section image in a predetermined posture does not always include a portion of interest and image features used as references, and hence it is not always easy to observe and diagnose the portion of interest based on the image features of the portion of interest and its peripheral portion. In addition, Japanese Patent Laid-Open Nos. 2005-169070 and 2011-125568 give no consideration to a technique of generating a cross section image including a designation point indicating a portion of interest designated in three-dimensional image data and feature points indicating known image features.
The present invention provides an image processing technique capable of generating a cross section image including a designation point indicating a portion of interest in three-dimensional image data and an image feature as a reference.
In addition, the present invention provides an image processing technique capable of generating a plurality of cross section images including designation points indicating portions of interest in three-dimensional image data and feature points to be sequentially selected.